evidence-auditor

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • Prompt Injection (LOW): The skill is susceptible to indirect prompt injection as it ingests untrusted data from document claims.\n
  • Ingestion points: Reads manuscript claims from output/CLAIMS.md.\n
  • Boundary markers: Absent. The skill lacks explicit markers or instructions to isolate analyzed content from its core logic.\n
  • Capability inventory: The skill is restricted to reading local files and writing findings to output/MISSING_EVIDENCE.md. It has no network, shell, or code execution capabilities.\n
  • Sanitization: No input validation or escaping is specified for the data processed.\n- No Code (SAFE): Analysis confirms this skill contains only natural language instructions and no executable scripts, packages, or binary files.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:07 PM